ME 380

Chapter 4 HW

February 27, 2012

Chapter 4 HW Solution
Review Questions.
1. Name the performance specification for first order systems. Time constant .
2. What does the performance specification for a first order system tell us? How fast the system responds.
5. The imaginary part of a pole generates what part of the response? The un-decaying sinusoidal part.
6. The real part of a pole generates what part of the response? The decay envelope.
8. If a pole is moved with a constant imaginary part, what will the responses have in common?
Oscillation frequency.
9. If a pole is moved with a constant real part, what will the responses have in common? Decay envelope.
10. If a pole is moved along a radial line extending from the origin, what will the responses have in
common? Damping ratio (and % overshoot).
13. What pole locations characterize (1) the underdamped system, (2) the overdamped system, and
(3) the critically damped system?
1. Complex conjugate pole locations.
2. Real (and separate) pole locations.
3. Real identical pole locations.
14. Name two conditions under which the response generated by a pole can be neglected.
1. The pole is far to the left in the s-plane compared with the other poles.
2. There is a zero very near to the pole.

Problems.
Problem 2(a). This is a 1st order system with a time constant of 1/5 second (or 0.2 second). It also has a DC gain
of 1 (just let s = 0 in the transfer function).

The input shown is a unit step; if we let the transfer function be called G(s), the output is input transfer function.
The resulting response function C(s) is #9 in my Laplace transform table, or you can expand the result in partial
fractions,
C(s) =

5
1
1
1
G(s) =
=
s
s(s + 5)
s s+5

(1)

Either way, the resulting response c(t) is

c(t) = 1 e5t

(2)

The time constant, rise time, and 2% settling time are:

= 1/5 sec
Tr = 2.2 = 0.44 sec
2% Ts = 4 = 0.8 sec
1

(3)

ME 380

Chapter 4 HW

February 27, 2012

Problem 3(a). Same system as above, but use MATLAB step

function to find step response. I did something like:
numG = [0 5]; % Define TF numerator
denG = [1 5]; % Define TF denominator
G = tf(num,den); % Define transfer function
[y,t] = step(G); % Find step response
plot(t,y); % Plot step response

0.8

The step response is shown in Figure 1 at right. At 0.2 seconds

the response is 63% of the way to the final value. Hopefully
the rise time and settling time are also about right.

Response c(t)

>>
>>
>>
>>
>>

0.6
0.4
0.2
0
0

0.2

Problem 8. (b) The TF is

T (s) =

0.4
0.6
Time (sec)

0.8

Figure 1: Step response

of first-order system with
= 0.2 sec.

5
= poles at s = 3, 6
(s + 3)(s + 6)

and the poles are shown below.

The general form of the step response will be

y(t) = A + Be3t + Ce6t

(4)

and it will be OVERDAMPED.

(d) This TF is
T (s) =

20
= poles at s = 3 j11.619
s2 + 6s + 144

and the poles are shown below.

The general form of the step response will be

y(t) = A + Be3t cos(11.619t + )

(5)

and it will be UNDERDAMPED.

Problem 9. To find the poles of

T (s) =

s4

s2 + 2s + 2
+ 6s3 + 4s2 + 7s + 2

one way is to use the MATLAB roots function:

2

(6)

ME 380

Chapter 4 HW

February 27, 2012

>> roots([1 6 4 7 2])

ans = -5.4917
-0.0955 + 1.0671i
-0.0955 - 1.0671i
-0.3173
So the poles of the given transfer function are:
s = 0.0955 j1.0671, 0.3173, 5.4917

(7)

Note that poles (roots) always occur as real numbers or complex conjugates. This is why all systems are made up
of first and second-order subsystems.

ME 380

Chapter 4 HW

February 27, 2012

Problem 18. The standard form of a second-order transfer function denominator is

s2 + 2n s + n2
By equating coefficients and solving for damping ratio and (undamped) natural frequency n , we get:
(b) (s + 3)(s + 6) = s2 + 9s + 18 from which we find n =
(d) s2 + 6s + 144 from which we find n =

18 = 4.24 rad/s, = 1.06 (overdamped)

144 = 12 rad/s, = 0.25 (underdamped)

Problem 20(c). Similar approach to the previous problem. The transfer function now is
T (s) =

1.05 107
K
= 2
2
3
7
s + 1.6 10 s + 1.05 10
s + 2n s + n2

(8)

The natural frequency is

n =

1.05 107 = 3, 240 rad/s (516 Hz)

(9)

The damping ratio is

2n = 1.6 103 = =

1.6 103
= 0.247
2n

(10)

Settling time, peak time, rise time, and % overshoot: these are all functions of and n .
4
= 0.005 sec
n

p
= 0.001 sec
Peak time Tp =
=
d
n 1 2
1.27
Rise time Tr
(Fig. 4.16) = 0.00039 sec
n
!

% Overshoot = exp p
100 = 45%
1 2

Settling time Ts =

(11)
(12)
(13)
(14)

Problem 21(c). The unit step response for the system of Problem 20(c) is shown in Figure 2 on the next page,
with the response characteristics indicated. I got them all from the response data rather than the expressions of
Problem 20(c).

ME 380

Chapter 4 HW

February 27, 2012

1.5
44.9% overshoot

Response

1
0.9

peak time 0.001 sec

0.5

settling time 0.005 sec

0.1

0
0

rise time 0.00039 sec

3
4
Time (sec)

7
3

x 10

Figure 2: Unit step response of Problem 21(c) using MATLAB. All

response characteristics obtained from data, not analytical expressions. They seem to agree quite well.

Problem 23. For the following second-order response specs, find the corresponding pole locations.
4
(a) Overshoot of 12% means = 0.55, and Ts =
= 0.6 sec means n = 6.67, so n = 12.1 rad/s. So the pole
n
location is
p
s = n jn 1 2 6.65 + j10.1
(b) Overshoot of 10% means = 0.6, and Tp =
the pole location is

= 5 sec means d = 0.628 sec, and n = d / 1 2 = 0.78. So

d

s = n jd 0.47 j0.628

(15)

= 7 sec means n = 0.57. Peak time Tp =

= 3 sec means d 1.05 rad/s. So like
n
d
part (b), the pole location is
(c) Settling time Ts =

s = n jd 0.57 j1.05

(16)

Problem 29(c). From Figure P4.9(c), the step response has 40% overshoot, hence the damping ratio 0.3.
The peakp
time Tp is about 4 sec, so the damped frequency d 0.78 rad/s. Then the undamped natural frequency
n = d / 1 2 0.82 rad/s. Finally, the DC gain is 1. So the transfer function is
0.67
n2
= 2
s2 + 2n s + n2
s + 0.49s + 0.67

ME 380

Chapter 4 HW

February 27, 2012

Problem 30. This is on pole-zero cancellation. As I indicated, this problem was poorly posed!! To change the
response functions into transfer functions, simply remove the s from the denominator of each function (this is
removing the unit step input). Then the C(s) becomes G(s) (a better letter to use for a transfer function).
I decided I just wanted for you to plot the unit step response of each original system plus that of the cancelled system
on the same plot (four plots: two responses on each plot). Adjust the numerator coefficient of the cancelled system
so the DC Gain of the cancelled system is the same as the original. The four TFs (original on left; cancelled on
right) are:
(a)
(b)
(c)
(d)

s+3
s+3
=
,
2
(s + 2)(s + 3s + 10)
(s + 2)(s + 1.5 j2.8)
s + 2.5
s+3
G(s) =
=
,
(s + 2)(s2 + 4s + 20)
(s + 2)(s + 2 j4)
s + 2.1
s+3
G(s) =
=
,
2
(s + 2)(s + 2s + 5)
(s + 2)(s + 0.5 j2.2)
s + 2.01
s+3
G(s) =
=
,
2
(s + 2)(s + 5s + 20)
(s + 2)(s + 2.5 j3.7)

1.5
+ 3s + 10
1.25
Gc (s) = 2
s + 4s + 20
1.05
Gc (s) = 2
s + 2s + 5
1.005
Gc (s) = 2
s + 5s + 20
Gc (s) =

G(s) =

s2

Note that I showed the complex pole locations for the quadratic polynomial in the second version of the original
transfer function.

0.18

0.08

0.16

0.07

0.14

0.06

0.12
Amplitude

Amplitude

0.05

0.1
Before cancellation
After cancellation

0.08

0.04

0.04

0.02

0.02

0.01

0
0

0.5

1.5

2.5

Before cancellation
After cancellation

0.03

0.06

3.5

0
0

0.5

Time (seconds)

(a) Difference in response; probably shouldnt cancel.

0.06

0.3

0.05

0.25

2.5

0.04

0.2

Amplitude

Amplitude

(b) A little less difference, but still some.

0.35

0.03

Before cancellation
After cancellation

0.15
Before cancellation
After cancellation

0.1

0.02

0.01

0.05
0
0

1.5
Time (seconds)

10

0
0

12

0.5

1.5

2.5

Time (seconds)

Time (seconds)

(c) Can definitely cancel here; almost no difference.

(d) Pole & zero completely cancel (almost).

Figure 3: The four step responses for Problem 30, showing situations where the pole and zero dont cancel, and when they do.

ME 380

Chapter 4 HW

February 27, 2012

DC Motor Problem. Use the system of part (c) in the Chapter 3 HW assignment, and find transfer function G(s),
where
G(s) =

L (s) rad
Ea (s) V

Plot the response of L (rad) to a 10V step input in motor voltage ea . Use MATLAB, and plot for 0.1 second.

Solution. From my notes, the transfer function from motor armature voltage ea (t) to load angular position L (t) is:
L (s)
=
Ea (s)

Kt
nR
 a Jt 
bt
s s+
Jt

(17)

where
n = gear ratio; in denominator of numerator to convert to L
JL
Jt = Jm + 2 (total inertia seen by motor)
n
bL
Kt Kb
bt = beq + 2 (and beq = bm +
)
n
Ra
The response to a 10V step input is shown below in Figure 3. Note that in 0.1 second the load moves about 0.8 rad
(45 ), which is the same as we saw in the Chapter 3 HW state-space model.

0.8
0.7

Load position (rad)

0.6
0.5
0.4
0.3
0.2
0.1
0
0

0.02

0.04

0.06

0.08

Time (sec)
Figure 4: Response of motor to 10V step input. Note that in 0.1
second the load moves about 0.8 rad (45 ), the same as in the
state-space model of Chapter 3.

0.1